When building for AI-powered environments, design tools to be equally usable by humans and the AI model. An elegant, simple design for humans often translates directly into an effective tool for AI agents, simplifying development and promoting shared logic.

For tools designed for AI interaction, the ease with which an agent can use the product (AX) is as critical as the user experience (UX) for humans. This can be improved by directly asking the agent for feedback on how to make the product more ergonomic for it.

In this software paradigm, user actions (like button clicks) trigger prompts to a core AI agent rather than executing pre-written code. The application's behavior is emergent and flexible, defined by the agent's capabilities, not rigid, hard-coded rules.

The paradigm of software development is shifting. Srini Raghavan argues that products must now be built for two types of users: humans and AI agents. This requires creating interfaces, like CLIs and APIs, specifically designed for AI consumption through technologies like MCPs.

To enable a 'bring your own agent' model, applications must offer dual interfaces. A traditional UI for the human user, and a machine-controllable programming interface (MCP or API) for the AI agent. The key is that both interfaces must modify the same underlying state in real-time for seamless collaboration.

A new software paradigm, "agent-native architecture," treats AI as a core component, not an add-on. This progresses in levels: the agent can do any UI action, trigger any backend code, and finally, perform any developer task like writing and deploying new code, enabling user-driven app customization.

A truly "agent-native" product goes beyond an API. The product's AI should be aware of its internal components—like project knowledge or UI elements—and possess the inherent ability to modify them directly, rather than just instructing a human on the necessary steps.

Descript's design principle for its AI agent, Underlord, is that it can't do anything a human user can't, and vice versa. This frames the AI as a true collaborator within the existing product interface, not a separate entity with special powers.

Designing for AI is less about crafting pixel-perfect UIs in Figma and more about creating the underlying system or "harness." This involves enabling the agent to perform long-running tasks, verify its own work, and operate effectively within technical constraints, which is where the real design work lies.